mradermacher/falcon-40b-GGUF
Quantization vs tiiuae/falcon-40b. mradermacher/falcon-40b-GGUF is a GGUF quantization (Q8_0) of the same weights: 242 tensors match exactly and 182 within quantization error.
Tensors changed: Norms 0/242, MLP 0/120, Attention 0/60, Embeddings 0/1, lm_head 0/1